The long-term objective of the project is to improve the image resolution, image quality, and production cost of PET and to broaden PET's applications in clinical, research and molecular imaging in oncology. The development of a new type of PET is proposed. This PET has an ultrahigh image resolution (2.2-2.6 mm) over current clinical PET (4.5 mm) and a transformable configuration including: (a) a high-sensitivity brain/animal1 mode, (b) a whole-body tumor localization/staging mode, (c) a high-sensitivity breast mode and (d) an extra large radiation treatment planning mode. The production cost of the PET would be less than that of a typical commercial PET. The dedicated breast mode has 13-26 times higher coincidence-detection sensitivity than regular PET's, and an image resolution of 2.2 mm, for early detection of small breast tumors and multi-lesion diseases. For brain imaging, the brain mode has an image resolution of 2.2 mm, a 4x higher detection sensitivity with an extra large 21 cm axial-field-of-view for concurrent imaging of the brain and carotid arteries (for obtaining arterial input function in the image enabled by the ultrahigh resolution). The health application aims of the project are to improve the accuracy of cancer diagnosis and research by (a) early detection of primary, metastatic and recurrent tumors as small as 2-3 mm, (b) determining the response of cancer to treatment earlier, (c) quantifying more accurately the functions of tumors, (d) facilitate quantitative tracer-kinetic modeling for cancer research, and (e) 1-2 minute fast breast imaging to facilitate breast screening for high risk young women and breast exams. The high-sensitivity, ultrahigh resolution small mode would be useful for animal models, e.g., genetically engineered mice. This proposed PET will use (a) an ultrahigh resolution, lower-cost detector design developed by us under NIH support, (b) a high speed PET electronics (that significantly reduce image artifacts) developed with NIH support for the ultrahigh resolution detectors, and (c) the results of the latest transformable-PET concept studies supported by NIH. A scaled-down prototype PET has been developed to demonstrate (a) the ultrahigh resolution detector concept, and (b) the theoretical validity and engineering feasibility of the transformable PET concept. This application aims to harvest 8 years of technology development and NIH funding through 4 large RO1 grants, all targeted for developing the proposed ultrahigh resolution transformable PET. With this full-size PET, clinical/human tests can be performed to test the multifunctionality of a transformable PET, the usefulness of dedicated small PET's for brain, head and neck, and breast imaging, and the usefulness of ultrahigh resolution PET imaging.